<html><head><meta http-equiv="Content-Type" content="text/html; charset=us-ascii"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;" class=""><div class=""><blockquote type="cite" class=""><div dir="ltr" class=""><div class="gmail_quote"><blockquote class="gmail_quote" style="margin: 0px 0px 0px 0.8ex; border-left-width: 1px; border-left-style: solid; border-left-color: rgb(204, 204, 204); padding-left: 1ex;"><div dir="ltr" class=""><div class="">Is the "cml_cont" field of the CmmCall variant is really used in practice?</div></div></blockquote></div></div></blockquote></div><div class=""><br class=""></div><div class="">Seconding what Michal has already said, yes, the `cml_cont` field is used quite a bit in Cmm.</div><div class=""><br class=""></div><div class="">Any non-tail call in the program will have this field populated with information about where control-flow continues after the call returns.</div><div class=""><br class=""></div><div class="">This field is used for some important steps, such as Proc Point analysis ( CmmProcPoint.callProcPoints ), the result of which is used to layout the stack. Stack layout uses the information about what values are live in the continuation block of a non-tail call to determine what values need to be saved to the stack.</div><div class=""><br class=""></div><div class="">Thus, the control-flow graph structure of a CmmProc is defined in part by the `cml_cont` field (see the Cmm specific instance of Hoopl's NonLocal class in CmmNode, where it is used to indicate successors of a block).</div><div class=""><br class=""></div><div class=""><blockquote type="cite" class=""><div dir="ltr" class=""><div class="gmail_quote"><blockquote class="gmail_quote" style="margin: 0px 0px 0px 0.8ex; border-left-width: 1px; border-left-style: solid; border-left-color: rgb(204, 204, 204); padding-left: 1ex;"><div dir="ltr" class=""><div class="">I traversed the output of raw Cmm produced by ghc compiling the whole base package, but the value of cml_cont is always Nothing.</div></div></blockquote></div></div></blockquote><br class=""></div><div class="">Hrm, were you looking for "returns to" in that output?</div><div class=""><br class=""></div><div class="">~kavon</div><div class=""><br class=""></div><div><br class=""><blockquote type="cite" class=""><div class="">On Mar 18, 2018, at 8:52 AM, Michal Terepeta <<a href="mailto:michal.terepeta@gmail.com" class="">michal.terepeta@gmail.com</a>> wrote:</div><br class="Apple-interchange-newline"><div class=""><div dir="ltr" class=""><div class="gmail_quote"><div dir="ltr" class="">On Sun, Mar 18, 2018 at 6:38 AM Shao, Cheng <<a href="mailto:cheng.shao@tweag.io" class="">cheng.shao@tweag.io</a>> wrote:<br class=""></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr" class=""><div class="">Hi all,</div><div class=""><br class=""></div><div class="">Is the "cml_cont" field of the CmmCall variant is really used in practice? I traversed the output of raw Cmm produced by ghc compiling the whole base package, but the value of cml_cont is always Nothing.<br class=""></div><div class=""><br class=""></div><div class="">Regards,</div><div class="">Shao Cheng</div></div></blockquote><div class=""><br class=""></div><div class=""><br class=""></div><div class="">Hi,</div><div class=""><br class=""></div><div class="">I'm not a GHC expert, so please don't trust everything I say ;)</div><div class=""><br class=""></div><div class="">That being said, I think `cml_cont` is used a lot. If you look at the</div><div class="">`compiler/codeGen` directory (that's what turns STG to cmm), you'll</div><div class="">see that `MkGraph.mkCallReturnsTo` is called a few times. That's the</div><div class="">function that will construct a `CmmCall` with the continuation block.</div><div class=""><br class=""></div><div class="">When dumping cmm, you'll often see all those `returns to` notes. For</div><div class="">instance, compiling:</div><div class=""><br class=""></div><div class="">```</div><div class="">foo :: Int -> Int</div><div class="">foo x =</div><div class=""> case x of</div><div class=""> 42 -> 111111</div><div class=""> _ -> 000000</div><div class="">```</div><div class=""><br class=""></div><div class="">results in:</div><div class=""><br class=""></div><div class="">```</div><div class=""> [...]</div><div class=""> c2cN: // global</div><div class=""> I64[Sp - 8] = c2cI;</div><div class=""> R1 = R2;</div><div class=""> Sp = Sp - 8;</div><div class=""> if (R1 & 7 != 0) goto c2cI; else goto c2cJ;</div><div class=""><br class=""></div><div class=""> // Evaluate the parameter.</div><div class=""> c2cJ: // global</div><div class=""> call (I64[R1])(R1) returns to c2cI, args: 8, res: 8, upd: 8;</div><div class=""> // ^^^^^^^^^^^^^^^</div><div class=""> // this specifies the continuation block</div><div class=""> // see also PprCmm.pprNode</div><div class=""><br class=""></div><div class=""> // Now check if it's 42.</div><div class=""> c2cI: // global</div><div class=""> if (I64[R1 + 7] == 42) goto c2cU; else goto c2cT;</div><div class=""> c2cU: // global</div><div class=""> [...]</div><div class="">```</div><div class=""><br class=""></div><div class="">As far as I understand it, this allows the code above to jump to the</div><div class="">`x` closure (to evalutae it), and have the closure jump back to the</div><div class="">continuation block (note that it's address is stored before we jump to</div><div class="">closure). AFAICS this particular code is created by</div><div class="">`StgCmmExpr.emitEnter`.</div><div class=""><br class=""></div><div class="">Hope this helps!</div><div class=""><br class=""></div><div class="">- Michal</div><div class=""> </div></div></div>
_______________________________________________<br class="">ghc-devs mailing list<br class=""><a href="mailto:ghc-devs@haskell.org" class="">ghc-devs@haskell.org</a><br class="">http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs<br class=""></div></blockquote></div><br class=""></body></html>